Seventeen bacterial species of Campylobacter have been identified to date. Cultivation test is commonly used to identify Campylobacter bacterial species. However, the test requires complex and substantial effort because some bacterial species are difficult to identify based on their biochemical properties alone. Also, the bacteria are microaerophilic and depending on the bacterial species, some need to be cultured at different temperatures. Furthermore, the cultivation test for Campylobacter bacteria including isolation and identification usually takes a long time (seven to ten days).
More simple and rapid methods for identifying various species of Campylobacter bacteria are expected to be developed, because there is an increasing trend in both the Campylobacter infection rate and number of patients (“Food poisoning outbreak for each causative agent”, the Ministry of Health, Labor and Welfare of the Japan).
It is difficult to rapidly identify Campylobacter bacterial species based on their biochemical properties, and some of Campylobacter species often cannot be distinguished based on their biochemical properties because of their close resemblance. For example, Campylobacter jejuni (hereinafter referred to as “C. jejuni”) and Campylobacter coli (hereinafter referred to as “C. coli”) are problematic because they are distinguished based on the presence of hippuricase activity, and when the enzyme activity is low, C. jejuni is falsely identified as C. coli. For this reason, PCR methods for detecting the presence of the hippuricase gene have been used in actual tests. In recent years, 16S rRNA gene analysis is frequently used as a method for identifying bacterial species at the gene level. However, C. jejuni and C. coli are highly homologous to each other, and thus often cannot be distinguished from each other by the 16S rRNA gene analysis.
To date, C. jejuni and C. coli account for about 94% and 4% of Campylobacter bacteria isolated from diarrhea patients, respectively. That is, the two bacterial species comprise the majority of Campylobacter bacteria. Thus, in most cases, test for Campylobacter bacteria in clinical practice only covers C. jejuni and C. coli which are specified as food poisoning bacteria. Furthermore, selection media commonly used in the test have been developed for mainly C. jejuni and C. coli, and the culture is generally carried out at 42° C. On the other hand, this bacterial isolation method is not suitable for bacterial species other than C. jejuni and C. coli because isolation of other bacterial species is less frequent. Specifically, depending on the selection medium or culture conditions used, sometimes bacterial species other than C. jejuni and C. coli cannot be isolated due to differences in the antibiotic sensitivity or optimal culture temperature among bacterial species belonging to the genus Campylobacter. That is, it is hard to say that the test covers Campylobacter fetus (hereinafter abbreviated as “C. fetus”) which has different temperature-sensitive property, or other Campylobacter bacteria.
Meanwhile, bacterial species other than C. jejuni and C. coli are also distributed in the gastrointestinal tract of pets, domestic and wild animals or such, and thus the chance of human infection is thought to be high as with C. jejuni and C. coli. A mass outbreak of food poisoning caused by C. fetus occurred in Osaka in 2005. Infection with C. fetus causes not only gastroenteritis such as diarrhea but also other severe symptoms such as sepsis and meningitis in human. Furthermore, infection with C. fetus can result in infertility, miscarriage, or the like in animals such as cattle. In addition to C. jejuni, C. coli, and C. fetus, the three bacterial species, Campylobacter lari (hereinafter abbreviated as “C. lari”), Campylobacter upsaliensis (hereinafter abbreviated as “C. upsaliensis”), and Campylobacter hyointestinalis (hereinafter abbreviated as C. hyointestinalis”), are zoonotic bacteria that cause enteritis, sepsis, or such in human. Thus, it is important to improve the system for testing Campylobacter bacteria other than C. jejuni, C. coli, and C. fetus. 
The present inventors cultured, isolated, and identified Campylobacter bacteria according to the Cape Town protocol without using antibiotics. The result showed that about 1.3% of patients with diarrhea caused by Campylobacter bacteria were infected with C. hyointestinalis (Non-patent Document 1).
C. hyointestinalis was isolated as a causative bacterium of porcine proliferative enteritis. Furthermore, C. hyointestinalis has been occasionally isolated from human enteritis patients, suggesting its involvement in human pathology. Nevertheless, there is no established rapid diagnosis method for C. hyointestinalis. 
Thus, although the chance of potentially infecting human is highly suspected, there is no appropriate isolation/culture test method for Campylobacter bacteria other than C. jejuni and C. coli. 
To solve the above-described problems, the present inventors focused and conducted their academic research on the cytolethal distending toxin (CDT) of Campylobacter bacteria (Non-patent Documents 2 and 3), and developed a method for detecting Campylobacter bacteria using the cytolethal distending toxin genes (cdtA, cdtB, and cdtC) (Patent Document 1). However, this detection method only targets C. jejuni, C. coli, and/or C. fetus, and no appropriate method has been developed for detecting other Campylobacter bacteria including C. hyointestinalis. 
Prior art documents related to the present invention described herein are shown below.    [Patent Document 1] WO 2005/054472    [Non-patent Document 1] Lastovica A J. et al., Campylobacter, 2nd ed, 89-120 (2000)    [Non-patent Document 2] Asakura M. et al., Microbial Pathogenesis 42 (2007) 174-183    [Non-patent Document 3] Yamasaki S. et al., Toxin Reviews, 25: 61-88, 2006